PL300 redesign

[TomD]

I purchased a PL300 in need of repair. I thought the build quality was pretty marginal so I decided to redesign the board rather than repair what I had. I modified the driver and bias circuitry and then after doing some testing I changed the output devices to lateral MOSFETs.
The stock heat sink looks large but based on the temperature rise it's actually undersized for the 125 watt output rating. I can bias the lateral FETs at a lower quiescent current (70 mA/device) and the heat sinks don't get nearly so hot.
My breadboard testing worked out very well so I decided to make a circuit board. I laid the board out with KiCAD. It's a 4 layer board with 1 ounce copper power power and return planes.
I've used it at least 4 days a week for the last couple months and I'm very happy with how it turned out.
The stock transformer voltage drops off quite a bit under load but I was able to get 132 watts out with a THD of 0.005% at 20 KHz and 0.0015% at 2 KHz. That meets or exceeds the original ratings.

 

[laatsch55]

Loaded, both channels driven?

 

[TomD]

It was only one channel driven and loaded with an 8 ohm power resistor when running on the stock transformer. I have connected both channels to a higher current transformer and measured 200 watts RMS on both under load.

The stock transformer is fine for typical music peaks and I suspect it would be good for 150 watts on a transient with both channels driven. I'd have to get a different source to test that though.

 

[laatsch55]

The outputs should be able to handle that...what was rail voltage with the bigger transformer...?

 

[laatsch55]

Certainly plenty of space on that board...density looks wonderful..

 

[TomD]

The rail voltage was +/- 68V, and I'm using 22000 uF caps. The transformer is almost twice as big and the voltage doesn't drop off under full load like it does with the stock transformer.

The board size is the same as the stock board so it will mount the same way.